Modelling the Association and Prediction of Cardiovascular Disease Using Log-linear and Binary Logistic Regression Approaches.

Abstract

Cardiovascular diseases are group of disorders of the heart and blood vessels that constitute the leading cause of death globally. This research focused on determining the factors associated with the likelihood of developing cardiovascular disease, and examining the pattern of association and interaction among categorical variables responsible for cardiovascular disease in Northwestern Nigeria. Data on cardiovascular risk factors and symptoms from a cohort comprising healthcare personnel and patients over a period extending from August 15, 2024 to September 15, 2024 were collected. Risk factors admitted by respondents including High Blood Pressure (Hypertension) acknowledged by 21.1% of respondents, High Cholesterol recognized by 15.2% of participants, Smoking practiced by 12.4% of the cohort, Alcohol Consumption reported by 8.2% of respondents, Unhealthy Diet admitted by 11.8% of participants, Physical Inactivity identified by 12.1% of the study population, Obesity and Overweight prevalent in 10.3% of respondents, and Diabetes recognized by 8.8% of the cohort which provided valuable intuitions into the prevalence of risk factors in the Northwestern Nigerian context. Findings from logistic regression affirmed age, gender, diagnosis, symptoms, risk factors, and family history as significant predictors of cardiovascular diseases with p-values 0.000, 0.000, 0.004, 0.000, 0.000, 0.002, and 0.001 respectively. Log linear modelling revealed positive partial association between age and gender, gender and residing site, age, gender, residing site with cardiovascular disease with p-values 0.0000, 0.0000, 0.0000, 0.0000, 0.0000 and 0.0000 respectively, while age and residing site with 0.0640 was not impactful.